There is no one delivery system which is ideal for all proposed gene therapy protocols. This proposal examines two aspects which would greatly expand the use of retroviral vectors, in particular those based on murine retroviruses. Retroviral vectors have many benefits for introduction of genes, including an efficient, site-specific viral integration system. One limitation of murine retroviral vectors is in the ability to infect non- dividing cells. Recently, the lentivirus class of retroviruses has been shown to infect both dividing and non-dividing cells. Most recent success has been in the generation of HIV base vectors, which raise concern for their overall use gene therapy. In this application, the knowledge that other retroviruses have adapted for infection of quiescent cells will be applied to the generation of murine retroviral vectors which are capable of infecting non-dividing cells. The research proposes a wide based, systematic approach to generate murine retroviral vectors which can infect quiescent cells. Three general schemes are proposed. The first is the insertion of nuclear localization signals from multiple alternative import pathways into the viral integrase protein (IN). The second approach is the generation of IN fusion proteins with proteins known to be nuclear or involved in the import process. Finally, alternatives MuLV the proteins which are associated with the preintegrative complex will be modified. The second specific aim addresses the potential to target integration within the host genome. Although viral integration occurs site-specifically with respect to the viral genome, it occurs near randomly in the target DNA. Preliminary experiments are proposed to test a novel mechanism of targeting viral integration, using a protein complementation system.